[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Gnash-commit] [SCM] Gnash branch, master, updated. release_0_8_9_st
From: |
Sandro Santilli |
Subject: |
Re: [Gnash-commit] [SCM] Gnash branch, master, updated. release_0_8_9_start-422-gf615871 |
Date: |
Tue, 29 Mar 2011 19:30:44 +0200 |
On Tue, Mar 29, 2011 at 06:22:03PM +0200, Bastiaan Jacques wrote:
> On Tue, 29 Mar 2011, Bastiaan Jacques wrote:
>
> > Update to upstream Mozilla jemalloc (87f5bb93303c); no local changes.
>
> Please note the following:
>
> >+/*
> >+ * Use only one arena by default. Mozilla does not currently make
> >extensive
> >+ * use of concurrent allocation, so the increased fragmentation
> >associated with
> >+ * multiple arenas is not warranted.
> >+ */
> >+#define MOZ_MEMORY_NARENAS_DEFAULT_ONE
>
> This assumption may not hold for Gnash, which is something we should
> test, though my suspicion is that it does.
Gnash surely does allocations from multiple threads at the same time,
if that's the meaning of "concurrent allocation".
I've no idea about the implications of that.
--strk;
() Free GIS & Flash consultant/developer
/\ http://strk.keybit.net/services.html